Fujitsu and Tokyo Medical and Dental University harness the world’s fastest supercomputer and artificial intelligence technology for scientific discovery to shed light on drug resistance in cancer treatment
TOKYO, March 7, 2022 — (JCN Newswire) — Fujitsu and Tokyo Medical and Dental University (TMDU) today announced new technology that uses AI to uncover novel causal mechanisms of drug resistance in cancer treatments based on clinical data. Leveraging the world’s fastest supercomputer, “Fugaku”,(1), the new technology enables high-speed computation of 20,000 data variables in a single day and uncovers previously unknown causal relationships related to drug resistance in cancer cells of 1,000,000 billion cells different possibilities.
Fujitsu and TMDU applied this technology to gene expression level data(2) obtained from cancer cell lines to analyze drug resistance(3) against cancer drugs, and successfully extracted a new causal mechanism of a previously unknown gene that suggests a cause of lung cancer drug resistance. The new technology is expected to help accelerate drug discovery and achieve cancer therapies that are individualized for each patient.
The technology was developed under the theme of “elucidating the cause and diversity of cancer using large-scale data analysis and AI technology”, an initiative supported by TMDU , Kyoto University and Fujitsu within the framework of the Fugaku supercomputer(4) acceleration program .
Even if a patient is receiving targeted cancer therapy(5), the emergence of drug-resistant cancer cells poses a lifelong threat to complete remission. The mechanism by which some cancers become drug-resistant, however, remains unclear, and researchers continue to work on new analytical methods that shed light on how cells that have multiple driver mutations (6) acquire drug resistance. In drug development and clinical trials involving the repositioning of drugs(7), it is important to identify the patients for whom the drugs are expected to have an effect. However, the effectiveness of drugs can differ depending on the organ and the individual and variations in gene expression, and the number of models combining the expression levels of several genes exceeds 1,000 trillion(8). A full search of all 20,000 genes in the human genome would therefore take more than 4,000 years with a conventional computer and finding ways to speed up the process is a major challenge.
Newly developed technology
Fujitsu has implemented parallel conditional and causal algorithms to maximize computational performance with the Fugaku supercomputer to analyze the human genome in the time needed for practical research. By using Fujitsu’s “Wide Learning”(9) AI technology to extract potential gene combinations linked to the emergence of drug resistance based on statistical information, Fujitsu has developed a new technology that allows perform a full search in one day.
As a result of running data from the Dependency Map (DepMap)(10) portal using this technology on the Fugaku supercomputer, Fujitsu and TMDU were able to search the entire human genome for conditions and causation in a single day and determine the genes that cause resistance to drugs used to treat lung cancer(11).
Commentary by Professor Seiji Ogawa, Graduate School of Medicine, Kyoto University
Promising technologies like Fujitsu’s artificial intelligence technology for scientific discovery (“Wide Learning”) could one day contribute to the discovery of biomarkers, which represent a growing area of interest for drug development. The key to successful new drug development is identifying patients who may benefit from new drugs and conducting clinical trials. If the marker that predicts who will benefit from the drug is known, the cost of clinical trials can be significantly reduced and the likelihood of success in conducting individual clinical trials can be increased. From this point of view, pharmaceutical and other manufacturers should be very interested in this technology. The fact that it was implemented using Fugaku also raised expectations.
Going forward, Fujitsu and TMDU will conduct multi-layered, comprehensive analysis that combines various data, including time axis and location data, with the goal of accelerating medical research, including in the area of efficacy drugs, as well as to shed light on the causes of cancer. .
Fujitsu and TMDU will also collaborate on experimental research in the fields of drug discovery and medicine. TMDU will further use the technology developed in this research to promote research on strategies for incurable diseases such as cancer.
In addition to medical care, Fujitsu will use the new technology to solve problems in various fields, including marketing, system operations and manufacturing.
This research was conducted under the Ministry of Education, Culture, Sports, Science and Technology’s Fugaku Achievement Acceleration Program “Understanding the Origin and Diversity of Cancer through the large-scale data analysis and artificial intelligence technologies” (JPMXP 1020200102). Part of the research was conducted with the computing resources of the Fugaku Supercomputer (Issue #: hp 200138, hp 210167).
(1) “Fugaku” supercomputer:
A computer installed at RIKEN as a successor to the “K.” From June 2020 to November 3, it ranked first in 4 categories of supercomputer ranking for 4 consecutive years. The full operation started on March 9, 2021.
(2) Gene expression level:
Amount of RNA copied from DNA (the same nucleic acid as DNA synthesized by transcription using certain DNA sequences as templates).
(3) Drug Resistance:
Phenomenon in which the effect of a drug wanes as the drug is administered.
(4) Fugaku Supercomputer Achievement Acceleration Program:
Program launched in May 2020 by the Ministry of Education, Culture, Sports, Science and Technology with the aim of achieving rapid results.
(5) Targeted drug:
Drug designed to act only on the molecule (protein, gene, etc.) responsible for the disease.
(6) Transfer of driver:
A genetic mutation that directly causes the development or progression of cancer.
(7) Drug repositioning:
The application of existing drugs developed and approved for the treatment of one disease to the treatment of another disease.
(8) Over 1,000 trillion:
Even though the expression level of each gene is limited to a combination of 50 major genes known to be linked to cancer and the expression level of each gene is categorized into 2 categories (e.g., “high” gene expression or “weak”), the condition number is 2 to the 50th power, which exceeds 1,000 trillion.
(9) Broad Learning:
Official website “Hello, Wide Learning!”
(10) Map of dependencies (DepMap):
Data on the susceptibility and resistance of approximately 4,500 drugs to approximately 600 different cancer cell lines, provided by the American Broad Institute. Cancer cell line mutation information and expression data for all genes are included.
(11) Fujitsu and TMDU analyzed DepMap gene expression data from approximately 300 cancer cell lines, susceptibility and resistance data from Gefitinib (a molecularly targeted drug used to treat lung cancer and other types of cancer), and have extensively researched the conditions and mechanisms of cancer cells. lines that do not respond to Gefitinib. Fujitsu and TMDU identified conditions in which the expression levels of three transcription factors (genes that control the transcription of genes (RNA synthesis)), ZNF516, E2F6 and EMX1, were low. In lung cancer cell lines that meet these conditions, a mechanism triggered by the transcription factors SP7 and PRRX1 have been discovered as other potential causes of drug resistance in cancer cells (see reference image).
Fujitsu is the Japanese leader in information and communication technologies (ICT) offering a complete range of technological products, solutions and services. Around 126,000 Fujitsu people support customers in more than 100 countries. We use our experience and the power of ICT to shape the future of society together with our customers. Fujitsu Limited (TSE: 6702) reported consolidated revenue of 3.6 trillion yen (US$34 billion) for the fiscal year ended March 31, 2021. For more information, please visit www.fujitsu.com.
About Tokyo Medical and Dental University
Tokyo Medical and Dental University (TMDU) is Japan’s only comprehensive medical university and graduate school. It provided advanced medical treatment through a fusion of the medical and dental fields and strived to produce “professionals with knowledge and humanity”. TMDU contributes to human health and societal well-being by nurturing exceptional healthcare professionals with a human and global outlook.
Source: Fujitsu Ltd.
Copyright 2022 JCN Newswire. All rights reserved.